Depth cameras have attracted more and more attention in the field of machine vision, since they possess characteristics that the images shot by a depth camera can provide users with abundant information regarding the position between shot objects. For example, depth cameras are applied in the field of video monitoring, motion sensing game and the like.
The practical application of depth cameras involves aspects such as positioning, identifying, and tracking of a target person, which can usually render better effects within a large field of view, regardless of in the field of video monitoring or motion sensing game. However, for the existing depth images taken by a single depth camera, the field of view thereof is generally small, which can hardly meet the requirements of practical applications. Therefore, it is necessary to provide a method for synthesizing depth images shot by a plurality of depth cameras into a depth image with a large field of view.